Professional Documents
Culture Documents
Corbell A 2015
Corbell A 2015
Title: Audit Firm Rotation, Audit Fees and Audit Quality: The
Experience of Italian Public Companies
Author: Silvano Corbella Cristina Florio Giorgio Gotti Stacy
A. Mastrolia
PII:
DOI:
Reference:
S1061-9518(15)00024-5
http://dx.doi.org/doi:10.1016/j.intaccaudtax.2015.10.003
ACCAUD 204
To appear in:
Received date:
Revised date:
Accepted date:
17-10-2013
28-9-2015
1-10-2015
Please cite this article as: Corbella, S., Florio, C., Gotti, G., and Mastrolia, S.
A.,Audit Firm Rotation, Audit Fees and Audit Quality: The Experience of Italian
Public Companies, Journal of the Chinese Institute of Chemical Engineers (2015),
http://dx.doi.org/10.1016/j.intaccaudtax.2015.10.003
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
ip
t
cr
us
an
ABSTRACT
Ac
ce
p
te
This paper examines some of the costs and benefits associated with audit firm rotation using data
from Italy, where mandatory audit firm rotation has been in place since 1975. Previous studies in
this area did not find consistent evidence of an association between audit quality and voluntary
or mandatory audit firm rotation. A recent paper, examining Italian public companies audited by
a Big 4 audit firm, uses proprietary data and finds no statistically significant association between
audit firm rotation and audit quality. In this study, we hand-collect publicly available data for a
larger sample of Italian public companies audited by a Big 4 and non-Big 4 audit firm (1,583
firm-year observations) over a longer time horizon (1998-2011). We find that audit quality,
proxied by two different measures of earnings management, improves following audit firm
rotation for companies audited by a non-Big 4 audit firm. Additionally, we examine whether
higher audit fees are associated with audit firm rotation. Our results indicate that following audit
firm rotation, the total amount of fees paid to the auditor was lower for companies audited by a
Big 4 and unchanged for companies audited by a non-Big 4 audit firm. The results of this study
should be of interest to European and U.S. legislators who are currently, or have recently,
considered implementing mandatory audit firm rotation in order to improve financial reporting
quality.
Keywords: mandatory audit firm rotation, audit quality, audit fees, audit partner rotation, audit
market regulation.
Acknowledgement: We gratefully acknowledge the helpful comments and suggestions of the anonymous reviewers,
discussants, and workshop participants at the 2010 European Accounting Association Annual Meeting, at the 2010
European Institute for Advances Studies in Management (EIASM) Workshop on Accounting and Regulation, Siena
(Italy), 2011 American Accounting Association International Accounting Sections Midyear Meeting, Tampa (FL),
Page 1 of 49
Ac
ce
p
te
an
us
cr
ip
t
and 2012 European Institute for Advances Studies in Management (EIASM) 4th Workshop on Audit Quality, Santa
Margherita Ligure (Italy). All errors remain our own responsibility.
Page 2 of 49
ABSTRACT
an
us
cr
ip
t
This paper examines some of the costs and benefits associated with audit firm rotation using data
from Italy, where mandatory audit firm rotation has been in place since 1975. Previous studies in
this area did not find consistent evidence of an association between audit quality and voluntary
or mandatory audit firm rotation. A recent paper, examining Italian public companies audited by
a Big 4 audit firm, uses proprietary data and finds no statistically significant association between
audit firm rotation and audit quality. In this study, we hand-collect publicly available data for a
larger sample of Italian public companies audited by a Big 4 and non-Big 4 audit firm (1,583
firm-year observations) over a longer time horizon (1998-2011). We find that audit quality,
proxied by two different measures of earnings management, improves following audit firm
rotation for companies audited by a non-Big 4 audit firm. Additionally, we examine whether
higher audit fees are associated with audit firm rotation. Our results indicate that following audit
firm rotation, the total amount of fees paid to the auditor was lower for companies audited by a
Big 4 and unchanged for companies audited by a non-Big 4 audit firm. The results of this study
should be of interest to European and U.S. legislators who are currently, or have recently,
considered implementing mandatory audit firm rotation in order to improve financial reporting
quality.
Ac
ce
p
te
Keywords: mandatory audit firm rotation, audit quality, audit fees, audit partner rotation, audit
market regulation.
Page 3 of 49
1. Introduction
Agency theory indicates that the separation of management (agent) from ownership
(stakeholder) leads to a moral hazard problem because the agent (management) may pursue his
ip
t
own self-interest at the expense of the principal (stakeholder) (Jensen and Meckling 1976). The
moral hazard problem is amplified by information asymmetry between the two parties:
cr
managers, who run the company, know more about the company and its future prospects than do
us
shareholders. One way to reduce the consequences and the costs associated with moral hazard is
to hire an external third party an independent public accounting firm to audit the books,
an
professional standards, as well as their independence and objectivity, their knowledge of the
te
clients business operations and industry, and the audit teams working relationship with the
client companys management. There are two primary schools of thought regarding long audit
Ac
ce
p
firm tenure. One school believes that audit firms with relatively longer tenure have greater
knowledge of the companys business and industry, thereby providing a higher quality and more
efficient audit (Geiger and Raghunandan 2002; Johnson et al. 2002; Myers et al. 2003; Carcello
and Nagy 2004). The other school believes that audit firms with relatively longer tenure provide
an increased likelihood of familiarity (or even friendships) forming between the audit staff
members and client staff members, an increased likelihood of a stale audit program, and a
decreased likelihood that the auditor will make decisions contrary to the prior year decisions,
thereby providing a lower quality and less efficient audit (Defond and Subramanyam 1998; Arel
Page 4 of 49
et al. 2005; Gates et al. 2007; Dao et al. 2008; Daniels and Booker 2009). Interestingly, this latter
school of thought is driven primarily by perceptions not empirical evidence.
In an effort to strengthen auditor independence, many countries have legislated
ip
t
limitations on the auditor-client relationship including: mandatory audit partner rotation, hiring
and firing of the audit firm by the audit committee rather than management, internal reviews of
cr
audit engagements, and external peer or regulated reviews of audit engagements. Additionally,
us
some countries, including the United States (U.S.) post-Sarbanes-Oxley Act of 2002 (SOX),
limit the types of services a public accounting firm can provide to its audit clients,1 and the type
an
SOX, Section 203, requires rotation of the partner on an audit engagement of a public
company every 5 years, but does not, currently, require audit firm rotation (Bradshaw and
Sloan). In 2003, the General Accounting Office (GAO) released the results of a study on the
te
potential effects of mandatory audit firm rotation. The GAO concluded that mandatory audit
firm rotation may not be the most efficient way to strengthen auditor independence and improve
Ac
ce
p
audit quality considering the additional financial costs and the loss of institutional knowledge of
the companys previous audit firm of record, as well as the current reforms being implemented.
(Public Accounting Firms: Required Study on the Potential Effects of Mandatory Audit Firm
Rotation 2003, p.1). The GAO also suggested a wait and see attitude until the other reforms
put in place by SOX were in effect for several years, thereby leaving open the possibility that
audit firm rotation would be considered again in the future (Public Accounting Firms: Required
Study on the Potential Effects of Mandatory Audit Firm Rotation 2003).
1
2
Page 5 of 49
In August, 2011, the Public Company Accounting Oversight Board (PCAOB) issued a
Concept Release (no. 39) on auditor independence, objectivity, and professional skepticism,
including consideration of mandatory audit firm rotation. The comment period originally expired
ip
t
in December, 2011, but was extended to April of 2012 in order to solicit more feedback. In total,
the PCAOB received 659 comment letters related to this concept release; most letters
cr
vehemently opposed mandatory audit firm rotation. During July, 2013, the Financial Services
us
Committee of the U.S. House of Representatives took the decision out of the hands of the
PCAOB by overwhelmingly passing a bill amending the Sarbanes-Oxley Act of 2002 to prohibit
an
the PCAOB from requiring public companies to use specific audit firms or requiring public
companies to change audit firms on a rotating basis; the bill also directs the GAO to revisit their
2003 study mentioned above. The bill next will be taken up by the Senate Committee on
Banking, Housing, and Urban Affairs. Interestingly, in April of 2013, the European Parliament's
te
Legal Affairs Committee took related action by approving a draft law that would require public
entities to rotate audit firms every 14 years (with a possible extension to 25 years if certain
Ac
ce
p
This paper adds to the existing literature regarding mandatory audit firm rotation and also
informs both the decision taken by authorities in the U.S. to end discussion of mandatory audit
firm rotation and the seemingly opposite decision taken by the authorities in the European
Parliament. We examine some of the costs and benefits associated with mandatory audit firm
rotation using data from a country, Italy, where mandatory audit firm rotation has been in place
since 1975. Italy is one of the very few countries in the world to mandate audit firm rotation and
is, therefore, a unique setting to examine this topic. Specifically, we test whether there is a
change in audit quality associated with both mandatory and voluntary audit firm rotation. We
6
Page 6 of 49
also test whether there is a change in total audit fees paid to the auditor when there is a
mandatory or voluntary audit firm rotation.
A recent study examining Italian public companies audited by a Big 4 audit firm with
ip
t
private data provided by the Big 4 audit firms (Cameran et al. 2015) found no statistically
significant association between audit firm rotation and audit quality. We first replicate the results
cr
of Cameran et al. (2015) using publicly available data and then extend our sample to include
us
Italian public companies audited by a non-Big 4 audit firm and to examine a longer time period.
Extensive empirical research has shown that earnings quality is different for companies audited
an
by one of the Big 4 audit firms versus companies audited by non-Big 4 audit firms. This body of
research has examined both U.S. companies (DeAngelo 1981; Khurana and Raman 2004) and
Overall, our results indicate that for companies audited by non-Big 4 audit firms, audit
te
firm rotation is associated with an increase in audit quality without the added cost of an increase
in audit fees. By contrast, for companies audited by Big 4 audit firms, audit firm rotation is not
Ac
ce
p
associated with an increase in audit quality but is associated with a decrease in audit fees; these
latter results confirm the findings in previous literature. This study makes several contributions
to the literature. First, it replicates and extends a recent study (Cameran et al. 2015) using
publicly available data that include not only public Italian firms audited by a Big 4 audit firm but
also companies audited by a non-Big 4 audit firm. Second, our study expands the existing
literature to examine voluntary and mandatory audit firm rotation for both companies with Big 4
and non-Big 4 audit firms. The results of this study indicate that, for non-Big 4 audit firms audit
quality improves following audit firm rotation. Third, this study provides evidence that following
audit firm rotation companies with Big 4 audit firms experience lower audit fees, while
7
Page 7 of 49
companies with non-Big 4 audit firms do not experience a change in audit fees. These results
should be interesting to policy setters and regulators in Italy, policy setters and regulators in
countries considering implementing mandatory audit firm rotation (the European Parliament and
ip
t
others), the U.S. House of Representatives, the GAO, and academic researchers.
We organize the rest of this paper as follows. Section 2 provides a description of the
cr
institutional background, literature review and hypotheses development. Section 3 describes the
us
sample selection procedures and data collection. Section 4 describes the research design, our
measure of audit quality and our test models. Section 5 reports our results. Section 6 describes
an
our sensitivity tests and Section 7 concludes the paper and identifies the limitations of the study.
2. Institutional Background, Literature Review, and Hypotheses Development
Italy first legislated mandatory audit firm rotation in 1975 and has since made five
te
Ac
ce
p
The first regulation on mandatory audit firm rotation3 provided for an audit firm tenure of
three years, renewable, if desired, for two additional three year terms. Before the end of the
three-year appointment, voluntary audit firm changes are possible under certain conditions. After
nine consecutive years, a new audit firm must be appointed and the original firm must wait a
minimum of five years (cooling-off period) before the original firm could be reappointed. This
first regulation was in force until 1997.
Page 8 of 49
After 1997, the original decree was partially modified to no longer explicitly identify the
length of the cooling-off period. 4 Due to the vagueness in the new legislation, audit firms
commonly interpreted that the original audit firm could be reappointed after only one three year
ip
t
The third change in legislation occurred in 20055, in response to the well-known financial
cr
scandals of Cirio and Parmalat. This law modified the Law of Finance (Testo Unico della
us
Finanza 1998) extending the audit firm term from three years to six years, decreasing the
number of audit firm term renewals from three to two, and explicitly introducing a three year
an
cooling-off period. Under this new law, each audit firms existing term was extended from three
to six years, and the maximum tenure was extended from nine to twelve years. This law also
regulated audit partner rotation, requiring that the partner in charge of the audit be replaced by
another partner after six consecutive years. A three year cooling-off period is provided during
te
which the same partner may not be responsible for the audit of a previous auditee and its
associates, even if he/she works for another audit firm.
Ac
ce
p
After just one year, in 2006, the Law of Finance was modified again to extend audit firm
tenure from six years to nine years, renewable after a three year cooling-off period. 6 A
provisional rule was also introduced providing that all unexpired audit firm terms as of the
effective date of the Legislative Decree, with a total audit firm tenure of less than nine years,
could be extended to nine years at the next shareholders meeting. No changes were made to the
provisions regarding audit partner rotation.
Page 9 of 49
In January 2010, the fifth legislative change became effective extending the audit partner
tenure from six to seven years.7 The audit firm term and cooling-off period were not changed.
Overall these legislative changes, intended to improve audit quality, have increased the
ip
t
audit tenure compared to the first regulation of 1975, implicitly indicating that the Italian
authorities believed that relatively longer tenured auditors were associated with better audit
cr
quality.
us
We detail below some of the previous literature related to audit firm rotation and auditor
an
tenure, please see Stefaniak et al. (2009) for a more thorough review of this literature. Several
U.S. studies examine the relation between audit quality and auditor tenure and (voluntary)
auditor change. In general, these studies do not support the claims that long-tenured auditors are
associated with lower audit quality, indicating that mandatory audit firm rotation may not
te
improve audit quality as intended. In these studies, researchers have used several proxies for
audit quality: audit opinions (Geiger and Raghunandan 2002), discretionary accruals (Johnson et
Ac
ce
p
al. 2002; Myers et al. 2003), total accruals (Myers et al. 2003), persistence of accruals (Johnson
et al. 2002), and alleged fraudulent financial reporting (Carcello and Nagy 2004).
Several recent archival studies also fail to support mandatory audit firm rotation. First,
Jenkins and Velury (2008) examine the relation between audit firm tenure and conservatism for
U.S. publicly listed firms. Using different measures of conservatism, these authors find a positive
association between conservatism and the length of audit firm tenure. Interestingly, they find an
increase in conservatism between short8 and medium9 tenure auditor-client relationships, and this
7
8
10
Page 10 of 49
higher level of conservatism does not deteriorate for long10 tenure relationships. These results
indicate that audit firm rotation may have an adverse effect on the conservatism of reported
earnings due to a short tenure (lower conservatism) condition being frequently repeated.
ip
t
In an international study, Jackson et al. (2008) examine the relation between audit firm
tenure and audit quality for Australian companies. These authors find that audit quality, proxied
cr
by the likelihood of issuing a going concern opinion, increases as tenure increases. However,
us
when audit quality is proxied by discretionary accruals, no difference is noted when tenure
increases. These results would seem to indicate that mandatory audit firm rotation would not be
an
versus new auditors. The results of these studies are mixed. The results in Gates et al. (2007)
indicate that MBA students demonstrate more confidence in a companys financial statements
te
after audit firm rotation. Other studies find that both the capital and debt markets value longerterm auditors more than new auditors (Mansi et al. 2004; Ghosh and Moon 2005). More recently,
Ac
ce
p
Kaplan and Mauldin (2008) perform two experiments designed to test whether non-professional
investors in the U.S. hold different independence-related perceptions for audit partner rotation
versus audit firm rotation. This study proxies independence-related perceptions with how much
of an income decreasing audit adjustment management would be willing to record. Kaplan and
Mauldin find no statistically significant difference in independence-related perceptions between
the two rotation conditions, indicating that, for non-professional investors, audit firm rotation
does not seem to be associated with a higher perception of independence than audit partner
10
11
Page 11 of 49
rotation. The mixed results of these studies seem to indicate the need for further empirical testing
in this area.
Mandatory Audit Firm Rotation Literature
ip
t
Mandatory audit firm rotation, over time, may actually preclude selection of the most
qualified audit firm. On the other hand, successor auditors can offer a fresh perspective to the
cr
audit of a company. Audit firm rotation offers two advantages over partner rotation: first, a new
us
partner from a new audit firm may be more willing to contradict judgments made by the
predecessor partner; second, in a mandatory audit firm rotation environment, each partner is
an
aware that his/her judgments will be reviewed by another audit firm in a predetermined period of
time. Presumably, either of these circumstances could lead to improved audit quality.
Several countries currently have mandatory audit firm rotation regulation. Italy has
required audit firm rotation since 1975, Brazil since 1999, and Singapore has required audit firm
te
rotation for local banks since 2002. Numerous other countries including Austria, Canada,
Greece, Spain, Slovakia, and Turkey previously required mandatory audit firm rotation and have
Ac
ce
p
since eliminated it due, in part, to increased audit costs (Raiborn et al. 2006; Johnson 2007).
Peter Wyman, the (then) head of professional affairs for PricewaterhouseCoopers, stated in a
2005 article There is clear evidence from Italy and the US that audit firm rotation increases
costs to business, creates problems with audit quality in the period immediately after the change
of audit firms, and leads to further consolidation of audit work amongst the largest audit firms
(Wyman 2005). At the time of Wymans article, Italy was, and still is, the only of the EU
Member States requiring mandatory audit firm rotation. Additionally, in Italy, the Bocconi
University Report indicated that, while audit firm rotation is associated with reduced audit
quality, it seems to improve public confidence in corporations (Arel et al. 2005).
12
Page 12 of 49
Limited empirical research has been performed to date in actual mandatory audit firm
rotation regimes; one such paper is Ruiz-Barbadillo et al. (2009) which examines audit firm
rotation in Spain.11 These authors find that auditors were less likely to issue going concern
ip
t
opinions to financially stressed clients during the mandatory audit firm rotation regime than in
the six years following the mandatory rotation regime. These results would seem to indicate that
cr
mandatory audit firm rotation was not associated with improved audit quality in Spain.
us
In a recent paper, Cameran et al. (2015) use proprietary data provided by the Big 4 audit
firms in Italy to examine the relation between audit effort (quality) and audit fees for clients of
an
Big 4 audit firms between 2006 and 2009. Their results indicate that for their sample companies,
audit fees in the final year of an engagement were higher than normal, which the authors
attribute to opportunistic pricing. Additionally, the new audit firm appeared to discount their
audit fee, even though more hours were incurred on the engagement, which the authors attribute
te
to low-balling. Lastly, these authors find that audit quality was lower in the first three years of
the audit engagement, as compared to the last years of the previous audit firms tenure. In this
Ac
ce
p
study, we use publicly available data and a longer time period to both replicate the results of
Cameran et al. (2015) for companies audited by Big 4 audit firms and to extend current research
to examine companies audited by non-Big 4 audit firms.
Overall, the current empirical studies examining audit quality in a mandatory audit firm
rotation environment is both limited and conflicting, indicating, we believe, a need for further
research in this area. This paper provides both additional empirical evidence of an association
between audit quality and mandatory audit firm rotation, and evidence of a relation between
audit fees and audit firm rotation in a mandatory audit firm rotation environment. Due to
11
13
Page 13 of 49
conflicting results of studies examining audit firm change and audit quality, we formulate our
first hypothesis in the null form:
Hypothesis 1: There is no association between a change in audit firm and audit quality.
ip
t
Previous literature indicates that higher risk clients will choose higher quality auditors
cr
(Datar et al. 1991) and it is reasonable that audit firms will charge higher fees to higher risk
us
clients (Feltham et al. 1991). Several empirical studies support the relation between higher
(lower) client riskiness, more (less) auditor effort and higher (lower) audit fees (O'Keefe et al.
an
1994; Pratt and Stice 1994; Simunic and Stein 1996; Johnstone and Bedard 2003). As higher risk
clients are also more likely to have higher earnings management (abnormal working capital
accruals), the above literature supports including an audit fees control variable when modeling
earnings management.
te
Both the audit firm and the company invest significant effort and time (cost) following a
change in audit firms. This impact will be even larger for consolidated entities that require
Ac
ce
p
statutory audits in many countries. In a mandatory audit firm rotation environment, these startup
costs are more likely to be spread over fewer years, increasing the overall cost of the audit
function for both the audit firm and the audit client.12 Extensive research has documented a
relation between audit firm change and audit fees (Ettredge and Greenberg 1990; Pearson and
Trompeter 1994; Deis and Giroux 1996; Simon and Francis 1988; Cameran et al. 2015; Zain
2013). A recent paper indicates that the relation between these two variables continues in the
post-SOX period (Huang et al. 2009). Due to the relation between audit fees and audit firm
12
A 2003 GAO study determined that the average auditor tenure in the Unites States for Fortune 1000 companies
was approximately 22 years.
14
Page 14 of 49
change, the above literature supports including audit fees as a control variable when modeling
auditor change and reporting quality to avoid a correlated omitted variable problem.
Our second hypothesis tests whether there is an association between total fees paid to the
ip
t
audit firm and audit firm rotation. Previous literature suggests that mandatory rotation might
increase the auditors fees because of the increased amount of time the new audit firm has to
cr
spend to audit a new company. However, the higher startup costs may be spread over fewer years
us
and might not impact the audit fees paid immediately after the mandatory audit firm change.
Hence, we also formulate our second hypothesis in the null form:
an
Hypothesis 2: There is no association between a change in audit firm and the total /
abnormal audit fees paid to the audit firm by the company.
te
listed on the Milan Stock Exchange (Mercato Telematico Azionario, Borsa Italianas Main
Market) as of December 31, 2011. We then extend our sample backwards, to 1998.13
Ac
ce
p
We manually collect data about audit firm and audit partner changes from corporate
annual audit reports14 and shareholders meeting reports available on the Borsa Italiana website
or directly from the companys website. We also manually collect audit fee data between 1998
and 2006 from shareholders meeting reports and, for fiscal years ending on or after July 1, 2007,
from annual reports.15 Finally, we download accounting data for the period 1997-201116 from the
13
For the sake of transparency, we underline that the number of companies included in the sample decreases as we
go back in time due to both a reduced number of listed companies and limited publicly available data.
14
In Italy, annual audit reports are signed by the audit partner.
15
New legislation required that companies include audit fees and non-audit fees in the annual report.
16
In order to calculate abnormal working capital accruals, we need accounting data for the one year prior to our first
test year, therefore we collect accounting data from 1997 to 2011.
15
Page 15 of 49
AIDA database, which contains financial information for both listed and unlisted companies in
Italy. Our final sample for our main test consists of 1,583 company-year observations.
4. Research Design
ip
t
We use abnormal working capital accruals as calculated in DeFond and Park (2001) and
cr
Francis and Wang (2008) as our primary measure of audit quality; we also use abnormal accruals
us
as calculated in Han and Wang (1998) as a sensitivity test to make sure that our results do not
depend on the audit quality measure selected. 17,18 Our primary proxy for audit quality (DeFond
an
and Park 2001; Francis and Wang 2008; Cameran et al. 2015) estimates abnormal working
capital accruals as the difference between actual working capital for the current year and the
level of working capital predicted by the previous years working capital to sales ratio for each
company year. The calculated abnormal working capital accruals are expected to reverse against
te
future earnings, thereby shifting profit between reporting periods. Abnormal working capital
17
Ac
ce
p
(1)
Our proxies for audit quality differ from the Jones-type abnormal accruals models usually adopted by accounting
literature (Jones 1991; Kothari et al. 2005). For international data, the number of observations in each industry and
in each country can be small, thereby reducing the reliability of the results of abnormal accruals calculated using the
Jones model (Wysocki 2004; Meuwissen et al. 2007; Francis and Wang 2008). Not surprisingly, for our Italian
companies, we do not have enough observations in each year/industry combination to adopt a Jones-type model to
calculate abnormal accruals.
18
We also tried to calculate another abnormal accruals proxy as used in Francis and Wang (2008),
but this calculation for abnormal accruals requires historical cost (gross value) of PP&E, while the Italian accounting
data from Aida only provides information about net value of PP&E, and only a few observations (no more than 15)
report the value of accumulated depreciation, hence making impossible an ex-post calculation of gross PP&E value.
16
Page 16 of 49
=
=
ip
t
Where:
AWCA
WC
cr
(AWCA) to proxy for audit quality. (See Appendix 1 for details on how we calculated abnormal
us
an
We test our first hypothesis by examining the coefficients from a regression of audit
quality (AWCA) on a binary variable Change equal to 1 for the year following a change in audit
firm, with the new audit firm signing and taking responsibility for the content of the financial
Ac
ce
p
(2)
te
We control for variables that previous literature has found significant in explaining levels
and changes in abnormal accruals.19 We control for SIZE, proxied by the natural log of net sales,
as larger companies tend to have a relatively lower level of accruals compared to smaller
companies (Behn et al. 2008). We control for cash flow from operations (CFO) as previous
research has indicated an inverse relation with accruals (Dechow 1994; Sloan 1996). To proxy
for the likelihood of financial distress, we control for leverage (LEV) and LOSS (a binary
19
All control variable calculations from AIDA data are detailed in Appendix 1.
17
Page 17 of 49
variable equal to 1 if the company incurred a loss in the previous period and zero otherwise) as
previous research has shown that either of these variables can proxy for the companys incentive
to use accruals to manage earnings in order to avoid violating debt covenants or other market
ip
t
reactions to negative income (DeFond and Jiambalvo 1994; Jaggi and Lee 2002; Behn et al.
2008). We control for the growth opportunities of a company, proxied by the year over year
cr
percent of sales growth (SALESGR), as literature has shown accruals are strongly correlated with
us
this measure (Carey and Simnett 2006; Behn et al. 2008). We also include return on assets (ROA)
in our model to control for the nondiscretionary component of abnormal accruals (Behn et al.
an
2008).
After running the regression model for the overall sample, we partition our sample into
two sub-samples: public companies audited by a Big 4 audit firm20 (BIG = 1) and public
companies audited by a non-Big 4 firm (BIG = 0, 2, 3). To partition the sample we use the
te
variable BIG: BIG = 1 if the companys audit firm is one of the largest four audit firms in the
country, 0 if the audit firm is not one of the Big 4 audit firms, 2 if the audit firm is Arthur
Ac
ce
p
Andersen, and 3 if we do not find information about the audit firm in the companys financial
report.21
We test whether the relation between audit firm change and earnings quality is different
when the change in audit firm is voluntary (Volchange) versus mandatory (Manchange).22 We
also examine whether there is a relation between audit partner change (Partchange) (without a
20
In Italy, the four largest auditors in the country include Reconta Ernst & Young (REY), PricewaterhouseCoopers
(PWC), Deloitte & Touche (D&T), and KPMG.
21
We assume that for all firms where BIG is coded as three there were no changes in auditor for the sample period,
thus biasing the results against finding evidence of an impact of auditor change on earnings quality.
22
More specifically, we replace our test variable Change with these other two variables: Volchange and Manchange.
Please see Table 4 for details.
18
Page 18 of 49
change in audit firm) and earnings quality and audit fees.23 Finally, we test whether there is a
difference in earnings quality after the change in audit firm between companies audited by a Big
4 audit firm and companies audited by a non-Big 4 firm. Using this methodology, we are able to
ip
t
first replicate the results reported in previous literature (Cameran et al. 2015) using our sample of
publicly available data that covers a longer time period to document the relation between audit
cr
firm rotation and earnings quality and audit fees for companies using Big 4 audit firms. Second,
us
we are able to extend existing literature by examining the relation between audit firm rotation
(mandatory, voluntary, and partner only) and earnings quality and audit fees for companies using
an
non-Big 4 audit firms. This second part of the analysis is new and can provide us with some
insight about whether the impact of mandatory/voluntary audit firm rotation on audit quality is
We also test whether there is a change in audit fees associated with audit firm rotation.
te
We use two measures of audit related fees: first, we measure audit fees as audit related fees
(3)
Ac
ce
p
Second, we calculate abnormal (excess) audit related fees (abnFees) as suggested in Hope et al.
(2009).
(4)
23
We code this variable for our first year of data as zero (no audit partner change) assuming there was no change in
the audit partner between the first year and the previous year.
19
Page 19 of 49
Specifically, we calculate abnormal audit related fees by regressing total audit related fees
(TOTFEE scaled by total sales for the year) over a broad set of explanatory variables (see
ip
t
equation (5) below) and then we use the residuals of the regression as a proxy for excess audit
fees (abnfees) ie., the portion of audit related fees not explained by the independent variables in
(5)
an
us
cr
the model.
We run models (3) and (4) for our entire sample and separately for Big 4 and other audit firms.
5. Results
Descriptive statistics are presented in Table 2: total assets for the sample companies
te
averages 4,110 million Euro (median 310 Million Euro). Net Income for the sample companies
Ac
ce
p
The correlation matrix presented in Table 3 shows a negative and significant association
between abnormal working capital accruals our first measure of earnings quality (AWCA) and
company size (SIZE) and the leverage variable (LEV); a positive and significant association
between abnormal working capital accruals (AWCA) and previous year loss (LOSS). Our second
measure of earnings quality, the abnormal accruals calculated as in Han and Wang (1998)
(ABHW), is also positively associated with previous year loss (LOSS) and negatively associated
with company size (SIZE) and returns on assets (ROA).
[Insert Table 3 about here]
20
Page 20 of 49
Multivariate Results
Multivariate analysis examines whether audit firm rotation (Change) is
positively/negatively associated with audit quality, as proxied by the absolute value of abnormal
ip
t
working capital accruals (AWCA) while controlling for those variables previous research has
shown to have an independent effect on abnormal accruals, including audit fees and audit partner
cr
rotation24. Table 4 presents these results. Column 1 presents results for our full sample; Columns
us
2 and 3 present results for Big 4 audit firms and non-Big 4 audit firms, respectively. To
determine if there is any difference between mandatory and voluntary audit firm rotation, we
an
introduce two binary variables to replace the variable Change: Volchange is equal to 1 if the
company voluntarily changes audit firm, zero otherwise; Manchange is equal to 1 if the company
mandatorily changes audit firm, zero otherwise. Column 4 presents the results of this model for
the full sample; Columns 5 and 6 present the results for Big 4 audit firms and non-Big 4 audit
te
firms, respectively. Finally we introduce a binary variable Partchange that takes the value of
1 for the year when the companies in our sample have a different audit partner signing the audit
Ac
ce
p
report, but there was NOT a change in audit firm. Column 7 presents the results of this model for
the full sample; Columns 8 and 9 present results for Big 4 audit firms and non-Big 4 audit firms,
respectively.
We find similar to the results from previous literature for Italian public companies
audited by a Big 4 audit firm (Cameran et al. 2015) no significant association between audit
firm rotation (voluntary and mandatory) and audit quality in the full sample and in the subsample of companies audited by a Big 4 audit firm (Column 1, 2, 4, 5, 7, and 8). However, when
24
21
Page 21 of 49
we limit the sample to public Italian companies audited by a non-Big 4 audit firm, we find
negative and significant coefficients for AWCA (Tables 4, Columns 3, 6, and 9). These results
indicate that audit quality improves in the year of the audit firm change (both mandatory and
ip
t
voluntary audit firm change) for Italian public companies audited by non-Big 4 audit firms25. We
also find a negative and significant relation between audit partner change and audit quality
cr
us
We use multivariate regression to test whether there is a change in the amount of audit
related fees (totrevscaled Table 5) or abnormal fees (abnfees Table 6) paid to the audit firm
an
related to the audit firm rotation. Table 5 shows that as documented in a recent paper (Cameran
et al. 2015) for Italian public companies audited by a Big 4 audit firm, there is a statistical
significant reduction in total audit fees paid to the audit firm in the first year after the audit firm
change: Big 4 auditors seem to discount their audit fees in an effort to compete for new clients.
te
However, our results also indicate that there is not such a change in audit fees for companies
audited by a non-Big 4 audit firm. Indeed there is no evidence of a significant coefficient for the
9).
Ac
ce
p
variable Change (Table 5, Column 3) or for Volchange or Manchange (Table 5, Column 6 and
Finally, we test whether abnormal audit fees are associated with a change in audit firm or
audit partner. Table 6 presents our results. Similar to the results in Table 5, we find evidence of a
reduction in abnormal audit fees for the sub-sample of companies audited by a Big 4 audit firm
25
In a sensitivity check, we test whether this change in audit quality persists in the years after the audit firm change.
We find that for non-Big 4 audit clients the improved earnings quality persists for up to two years after the audit
firm change.
22
Page 22 of 49
but no evidence of a reduction in abnormal audit fees for companies audited by non-Big 4 audit
firms (Table 6, Column 6 and 9).
[Insert Table 6 about here]
ip
t
Overall, our results indicate that for companies audited by non-Big 4 audit firms, audit
firm change is associated with an increase in audit quality without a change in audit fees.
cr
6. Sensitivity Checks
us
We conduct a number of additional analyses to validate the results of our main regression
models.
an
Companies audited by Arthur Andersen were included in our main analysis as BIG = 0. If
we perform the main analysis with Arthur Andersen identified as a Big 4 audit firm, our results
te
Our second proxy for audit quality (Han and Wang 1998) calculates abnormal accruals in
Ac
ce
p
a two-step calculation. In the first step, we calculate the residuals of a model estimating the
change in sales as a function of the change in working capital (while controlling for year fixed
effects).
Where:
WC
WC
S
S
=
=
=
(6)
23
Page 23 of 49
In the second step, we calculate abnormal accruals (our second proxy for audit quality)
using the residuals from this regression model divided by total sales revenue (ABHW).
ip
t
(7)
The control variables are the same as identified above in our main analysis.
cr
We partition our sample into two sub-samples: public companies that are audited by one
us
of the largest audit firms vs. public companies audited by a non-Big 4 firm, using the variable
BIG as defined above in our main analysis. We also control for audit partner rotation with a
an
binary variable (Partchange) as defined above in our main analysis. Results and inferences,
reported in Table 7, are qualitatively the same as the results we report in the main results section
of the manuscript.
te
6.3 Persistence of Audit Quality after the Audit Firm Change and Pre vs. Post Change
To test whether the difference in audit quality persists for the years after the change in
Ac
ce
p
audit firm, and to ensure that there is a change in audit quality before vs. after the audit firm
change, we introduce binary variables: Prechange, equal to 1 for the year before the change in
audit firm and zero otherwise, Postchange, equal to 1 for one year after the audit firm change
and zero otherwise, Post2change, equal to 1 for the second year after the audit firm change and
zero otherwise, Post3change, equal to 1 for the third year after the audit firm change and zero
otherwise. Table 8 reports our results: audit quality improves following a change in non-Big 4
audit firm, while the coefficient of the Prechange binary variable is not statistically different
from zero for companies audited by non-Big 4 audit firms and positive but only weakly
significant for companies audited by Big 4 audit firms. There is also strong evidence that the
24
Page 24 of 49
improvement in audit quality for companies audited by a non-Big 4 audit firm persists after the
audit firm change, as each of the post-change binary variables have negative and strongly
ip
t
significant coefficients.
cr
Following previous literature (among others, Cameran et al. (2015)), our main analysis
us
used the absolute value of abnormal accruals as the dependent variable. As a sensitivity test, we
also ran model (2) separately for positive (income increasing) and negative (income decreasing)
an
accruals. Table 9 reports our results for the signed accruals tests: for companies in the income
increasing accruals subsample, audit quality improves following a change in non-Big 4 audit
firm (voluntary change, mandatory change, and partner change); for companies in the income
decreasing accruals subsample, audit quality improves following a change in non-Big 4 audit
te
firm, but only for voluntary change and partner change. For companies audited by a Big 4 audit
firm, none of the coefficients were statistically significantly different from zero.
Ac
ce
p
To ensure that our results in particular our results for the non-Big 4 subsample are not
driven by one or two firms with extreme values, we winsorized (at 2%, both tails) the dependent
variable in our main model (AWCA). Since the subsample of companies audited by non-Big 4
firms includes 599 observations, this will exclude from our analysis the top and bottom 12 most
extreme observations. Our results and inferences do not change for either of the Big-4 or non-Big
4 subsamples.
25
Page 25 of 49
6.6 Statistical Significance of the Difference in Audit Quality between Big 4 and non-Big 4
Clients Following Audit Firm Change
In Table 4, we test the relation between abnormal accruals and multiple types of auditor
ip
t
change for Big 4 clients and non-Big 4 clients in separate models. We find results for non-Big 4
clients in the accrual model but we did not test specifically whether these results are significantly
cr
different between the two sub-samples. In order to test for statistical differences in the regression
us
coefficients between Big 4 and non-Big 4 clients, we run a model with both Big 4 and non-Big 4
clients, and add a binary variable nonbig4 equal to one for non-Big 4 clients, zero otherwise.
an
Then, we include this variable and its interactions with the other independent variables in the
te
(8)
Ac
ce
p
Table 10 reports our results: audit quality is higher for firms audited by non-Big 4 following a
change in audit firms.
Authorities in the U.S. and elsewhere continue to evaluate whether the benefits of audit
firm rotation outweigh the costs, and whether audit firm rotation might provide higher audit
quality as compared to individual partner rotation. This study examines this question in the
unique environment of a country, Italy, with a long history of mandatory audit firm rotation, with
the goal of informing the discussion related to mandatory audit firm rotation elsewhere in the
26
Page 26 of 49
world. Our sample includes all listed Italian companies, from 1998 to 2011, with the necessary
accounting and audit data to perform our calculations. Our results indicate that audit quality,
proxied by two measures of abnormal accruals, improves following audit firm rotation (both
ip
t
mandatory and voluntary) but only for companies audited by a non-Big 4 audit firm, even after
controlling for audit partner rotation and other variables previous research has shown to have an
cr
association with audit quality. Additionally, our results indicate that the total fees paid to the
us
audit firm do not change following mandatory audit firm rotation for companies audited by a
non-Big 4 audit firm.
an
statements are a joint product of the company management and the audit firm and, therefore, we
are not able to isolate the effects of a companys financial reporting quality from the audit quality
te
itself. As with all studies that use this measure, we cannot observe earnings quality directly,
however, we do use estimates calculated by two different models in this study.
Ac
ce
p
Subject to these limitations, this study makes several contributions to the literature. First,
it provides empirical evidence that audit quality improves following an audit firm rotation in
Italy (both mandatory and voluntary audit firm rotation) for companies audited by a non-Big 4
audit firm. Second, it provides some evidence that total fees paid to the audit firm and the
abnormal fees paid for audit services do not change following audit firm rotation for these
companies. As a result of these contributions, we believe the results of this study should be
interesting to policy setters and regulators in Italy, policy setters and regulators in countries
considering implementing mandatory audit firm rotation (the European Parliament and others),
the U.S. House of Representatives, the GAO, and academic researchers.
27
Page 27 of 49
List of References
Arel, B., R. G. Brody, and K. Pany. 2005. Audit Firm Rotation and Audit Quality. The CPA
Journal 75 (1):36-39.
ip
t
Behn, B., J.-H. Choi, and T. Kang. 2008. Audit Quality and Properties of Analyst Earnings
Forecasts. The Accounting Review 83 (2):327-349.
cr
Bradshaw, M. T., and R. G. Sloan. 2002. GAAP versus The Street: An Empirical Assessment of
Two Alternative Definitions of Earnings. Journal of Accounting Research 40 (1):41-66.
us
Cameran, M., J. R. Francis, A. Marra, and A. K. Pettinicchio. 2015. Are there adverse
consequences of mandatory auditor rotation? Evidence from the Italian experience.
Auditing: A Journal of Practice and Theory 34 (1):1-24.
an
Carcello, J. V., and A. L. Nagy. 2004. Audit Firm Tenure and Fraudulent Financial Reporting.
Auditing: A Journal of Practice & Theory 23 (2):55-70.
Carey, P., and R. Simnett. 2006. Audit Partner Tenure and Audit Quality. The Accounting
Review 81 (3):653-676.
Daniels, B. W., and Q. Booker. 2009. Loan Officers' Reactions to Audit Rotation. Journal of
Behavioral Studies in Business 1:1-13.
te
Dao, M., S. Mishra, and K. Raghunandan. 2008. Auditor tenure and shareholder ratification of
the auditor. Accounting Horizons 22 (3):297-314.
Ac
ce
p
Datar, S. M., G. A. Feltham, and J. S. Hughes. 1991. The role of audits and audit quality in
valuing new issues. Journal of Accounting & Economics 14:3-49.
DeAngelo, L. E. 1981. Auditor size and audit quality. Journal of Accounting & Economics 3
(3):183-199.
Dechow, P. M. 1994. Accounting earnings and cash flows as measures of firm performance :
The role of accounting accruals. Journal of Accounting and Economics 18 (1):3-42.
DeFond, M. L., and J. Jiambalvo. 1994. Debt covenant violation and manipulation of accruals.
Journal of Accounting and Economics 17 (1-2):145-176.
DeFond, M. L., and C. W. Park. 2001. The Reversal of Abnormal Accruals and the Market
Valuation of Earnings Surprises. The Accounting Review 76 (3):375-404.
Defond, M. L., and K. R. Subramanyam. 1998. Auditor changes and discretionary accruals.
Journal of Accounting & Economics 25 (1):35-68.
29
Page 28 of 49
Deis, D. R., Jr., and G. A. Giroux. 1996. The Effect of Auditor Changes on Audit Fees, Audit
Hours, and Audit Quality. Journal of Accounting and Public Policy 15:55-76.
Ettredge, M. L., and R. Greenberg. 1990. Determinants of Fee Cutting on Initial Audit
Engagements. Journal of Accounting Research 28 (1):198-210.
ip
t
Feltham, G. A., J. S. Hughes, and D. A. Simunic. 1991. Empirical assessment of the impact of
auditor quality on the valuation of new issues. Journal of Accounting & Economics
14:375-399.
cr
Francis, J. R., and D. Wang. 2008. The Joint Effect of Investor Protection and Big 4 Audits on
Earnings Quality around the World. Contemporary Accounting Research 25 (1):157-191.
us
Gates, S. K., D. J. Lowe, and P. M. J. Reckers. 2007. Restoring public confidence in capital
markets through auditor rotation. Managerial Auditing Journal 22 (1):5-17.
an
Geiger, M. A., and K. Raghunandan. 2002. Auditor tenure and audit reporting failures. Auditing
21 (1):67-78.
Ghosh, A., and D. Moon. 2005. Auditor Tenure and Perceptions of Audit Quality. The
Accounting Review 80 (2):585-612.
Han, J. C. Y., and S.-W. Wang. 1998. Political Costs and Earnings Management of Oil
Companies during the 1990 Persian Gulf Crisis. The Accounting Review 73 (1):103-117.
te
Hope, O.-K., K. Tony, W. B. Thomas, and Y. O. O. Yong Keun. 2009. Impact of Excess Auditor
Remuneration on the Cost of Equity Capital around the World. Journal of Accounting,
Auditing & Finance 24 (2):177-210.
Ac
ce
p
Huang, H.-W., K. Raghunandan, and D. V. Rama. 2009. Audit Fees for Initial Audit
Engagements Before and After SOX. Auditing: A Journal of Practice & Theory 28
(1):171-190.
Jackson, A. B., M. Moldrich, and P. Roebuck. 2008. Mandatory audit firm rotation and audit
quality. Managerial Auditing Journal 23 (5):420-437.
Jaggi, B., and P. Lee. 2002. Earnings Management Response to Debt Covenant Violations and
Debt Restructuring. Journal of Accounting, Auditing & Finance 17 (4):295-324.
Jenkins, D. S., and U. Velury. 2008. Does auditor tenure influence the reporting of conservative
earnings? Journal of Accounting and Public Policy 27:115-132.
Jensen, M. C., and W. H. Meckling. 1976. Theory of the firm: managerial behavior, agency costs
and ownership structure. Journal of Financial Economics 3 (4):305-360.
Johnson, M. 2007. does rotation work. Charter 78 (4):68-69.
30
Page 29 of 49
Johnson, V. E., I. K. Khurana, and J. K. Reynolds. 2002. Audit-firm tenure and the quality of
financial reports. Contemporary Accounting Research 19 (4):637-660.
Johnstone, K. M., and J. C. Bedard. 2003. Risk management in client acceptance decisions. The
Accounting Review 78 (4):1003-1025.
ip
t
cr
Kaplan, S. E., and E. G. Mauldin. 2008. Auditor rotation and the appearance of independence:
Evidence from non-professional investors. Journal of Accounting and Public Policy
27:177-192.
us
Khurana, I. K., and K. K. Raman. 2004. Litigation Risk and the Financial Reporting Credibility
of Big 4 versus Non-Big 4 Audits: Evidence from Anglo-American Countries.
Accounting Review 79 (2):473-495.
an
Kothari, S. P., A. J. Leone, and C. E. Wasley. 2005. Performance matched discretionary accrual
measures. Journal of Accounting and Economics 39 (1):163-197.
Mansi, S. A., W. F. Maxwell, and D. P. Miller. 2004. Does Auditor Quality and Tenure Matter to
Investors? Evidence from the Bond Market. Journal of Accounting Research 42 (4):755793.
te
Meuwissen, R., F. Moers, E. Peek, and A. Vanstraelen. 2007. Comparing Abnormal Accruals
Estimates Across Samples: An International Test: SSRN.
Ac
ce
p
Myers, J. N., L. A. Myers, and T. C. Omer. 2003. Exploring the term of the auditor-client
relationship and the quality of earnings: A case for mandatory auditor rotation? The
Accounting Review 78 (3):779-799.
O'Keefe, T. B., D. A. Simunic, and M. T. Stein. 1994. The production of audit services:
Evidence from a major public accounting firm. Journal of Accounting Research 32
(2):241-261.
Pearson, T., and G. Trompeter. 1994. Competition in the Market for Audit Services: The Effect
of Supplier OCncentration on Audit Fees. Contemporary Accounting Research 11
(1):115-135.
Pratt, J., and J. D. Stice. 1994. The effect of client characteristics on auditor litigation risk
judgements, required audit evidence, and recommended audit fees. The Accounting
Review 69 (October):639-656.
Government Accounting Office (GAO). 2003. Report to the Senate Committee on Banking,
Housing, and Urban Affairs and the House Committee on Financial Services. Public
Accounting Firms: Required Study on the Potential Effects of Mandatory Audit Firm
Rotation. November 21.
31
Page 30 of 49
Raiborn, C., C. A. Schorg, and M. Massoud. 2006. Should auditor rotation be mandatory? The
Journal of Corporate Accounting & Finance 17 (4):37-49.
Ruiz-Barbadillo, E., N. Gmez-Aguilar, and N. Carrera. 2009. Does Mandatory Audit Firm
Rotation Enhance Auditor Independence? Evidence from Spain. Auditing 28 (1):113-135.
ip
t
Simon, D. T., and J. R. Francis. 1988. The Effects of Auditor Change on Audit Fees: Tests of
Price Cutting and Price Recovery. The Accounting Review 63 (2):255-269.
cr
Simunic, D. A., and M. T. Stein. 1996. Impact of litigation risk on audit pricing: A review of the
economics and the evidence. Auditing: A Journal of Practice & Theory 15:119-134.
us
Sloan, R. G. 1996. Do Stock Prices Fully Reflect Information in Accruals and Cash Flows about
Future Earnings? The Accounting Review 71 (3):289-315.
an
Stefaniak, C. M., J. C. Robertson, and R. W. Houston. 2009. The Causes and Consequences of
Auditor Switching: A Review of the Literature. Journal of Accounting Literature 28:47121.
Testo Unico della Finanza. 1998.
Wyman, P. 2005. How do Europe's 8th directive and Sarbanes-Oxley compare? In gtnews.
Wysocki, P. D. 2004. Discussion of Ultimate Ownership, Income Management, and Legal and
Extra-Legal Institutions. Journal of Accounting Research 42 (2):463-474.
Ac
ce
p
te
Zain, M. M. 2013. Audit fees during initial engagement in Malaysia. Managerial Auditing
Journal 28 (8):735-754.
32
Page 31 of 49
ip
t
cr
AIDA ITEMS
1. ATTIVO CIRCOLANTE AL NETTO DEI CREDITI A
LUNGO TERMINE (CA)
Attivo circolante represents cash and assets expected to be converted
into cash and/or used in the production of revenue during next
operating cycle.
This item is the sum of:
1. Disponibilit liquide (cash and cash equivalents AIDA code
vc_1070)
2. Attivit finanziarie che non costituiscono immobilizzazioni (shortterm financial investments vc_1066)
3. Rimanenze (inventories vc_1045)
4. Crediti a breve termine (short-term receivables vc_1056)
As Attivo circolante may include long-term trade receivables, to
calculate CA we subtract such receivables from its total amount:
CA = vc_1071 vc_1075
2. ATTIVIT FINANZIARIE CHE NON COSTITUISCONO
IMMOBILIZZAZIONI E DISPONIBILIT LIQUIDE (FINA)
This item represents cash and funds convertible into cash within a
short period of time.
This item is a component of Attivo Circolante (variable CA above).
This item is the sum of:
1. Disponibilit liquide (cash and cash equivalents vc_1070)
2. Attivit finanziarie che non costituiscono immobilizzazioni (shortterm financial investments vc_1066)
FINA = vc_1070 + vc_1066
ce
pt
ed
M
an
us
APPENDIX 1
Variable Definitions
Ac
33
Page 32 of 49
ip
t
cr
3. DEBITI AL NETTO DEI DEBITI A LUNGO TERMINE
(LIABSHORT)
Debiti includes both short-term and long-term trade payables and
loans.
LIABSHORT is a component of Debiti and represents debts due
within one year. LIABSHORT is therefore the difference between
total trade payables and loans (vc_1118), and long-term trade
payables and loans (vc_1117).
LIABSHORT = vc_1118 vc_1117
4. DEBITI FINANZIARI A BREVE TERMINE (LIABFIN)
This item represents financial payables due within one year.
This item is the sum of:
1. Obbligazioni con scadenza entro lesercizio successivo (short-term
bonds vc_1090)
2. Obbligazioni convertibili con scadenza entro lesercizio successivo
(long-term bonds vc_1092)
3. Debiti verso soci con scadenza entro lesercizio successivo (shortterm borrowings from shareholders vc_1184)
4. Debiti verso banche con scadenza entro lesercizio successivo
(short-term bank loans vc_1094)
5. Debiti verso altri finanziatori con scadenza entro lesercizio
successivo (short-term borrowings from other financiers vc_1096)
LIABFIN = vc_1090 + vc_1092 + vc_11184 + vc_1094 + vc_1096
5. RICAVI NETTI DELLE VENDITE E DELLE PRESTAZIONI
(NETSALES)
This item represents gross sales reduced by cash discounts, sales
discounts and returns, and value-added taxes and allowances for
which credit is given to customers.
NETSALES = vc_1124
M
an
us
Ac
ce
pt
ed
34
Page 33 of 49
us
cr
ip
t
Ac
ce
pt
ed
M
an
35
Page 34 of 49
ip
t
cr
M
an
us
Ac
ce
pt
ed
Where
= Change in working capital accruals
Totrevscaled: the amount of audit service fees paid to the auditor scaled by total sales
abnFees: abnormal audit fees calculated as the excess audit fees amount as compared to projected audit fees
LOSS: binary variable equal to 1 if the company experienced a loss in the period, and 0 otherwise.
Change: binary variable equal to 1 for the year of the change in audit firm, with the new auditor signing and taking responsibility for the
content of the financial statements presented to investors, and 0 otherwise.
Volchange: binary variable equal to 1 for the year of a change in audit firm when the change is voluntary, and 0 otherwise
Manchange: binary variable equal to 1 for the year of a change in audit firm when the change is mandatory, and 0 otherwise
Partchange: binary variable equal to 1 for the year when there has been a partner change without an audit firm change, and 0 otherwise
Prechange: binary variable equal to 1 for the year before a change in audit firm, and 0 otherwise
Postchange: binary variable equal to 1 for the year after a change in audit firm, and 0 otherwise
Post2change: binary variable equal to 1 for the second year after a change in audit firm, and 0 otherwise
Post3change: binary variable equal to 1 for the third year after a change in audit firm, and 0 otherwise
SIZE: the natural log of net sales
LEV: the difference between total liabilities and stockholders equity over total assets
SALESGR: sales growth calculated as
ROA: return on assets from Aida
BIG: variable equal to 1 if the audit firm is one of the Big 4 audit firms in Italy (Reconta Ernest & Young (REY); PricewaterhouseCoopers
(PWC); DeloitteTouche Tohmatsu (D&T); KPMG), and zero if the audit firm is not one of the Big 4 audit firms in Italy. The variable is equal to
26
2 if the audit firm is Arthur Andersen, and equal to 3 if we do not find information about the auditor in the companys financial report
CFO: Cash flow from operations calculated as
26
We assume that for all firms where BIG is coded as three there were no changes in auditor during the sample period, thus biasing the results
against finding evidence of an impact of auditor change on earnings quality.
36
Page 35 of 49
TABLE 1
Legislation regarding mandatory audit firm rotation in Italy
1998 - 2004
2010 present**
an
2006 - 2009
us
cr
2005
Relevant Provisions*
Audit firm appointed for three years
Renewable for two additional three-year terms
Five years cooling-off period
Audit firm appointed for three years
Renewable for two additional three-year terms
Cooling-off period not explicit (often interpreted as one, three-year term)
Audit firm appointed for six years
Renewable for one additional six-year term
Three years cooling-off period
Audit partner replaced after six years, with a three years cooling-off
period
Audit firm appointed for nine years
Not immediately renewable
Three years cooling-off period
Audit partner replaced after six years, with a three years cooling-off period
Audit firm appointed for nine years
Not immediately renewable
Three years cooling-off period
Audit partner replaced after seven years, with a three years cooling-off
period
ip
t
Time
1975 - 1997
Ac
ce
p
te
38
Page 36 of 49
TABLE 2
Descriptive statistics
Panel A: Full Sample
Count
111,000,000
Change
160,000,000
4,371,000
(791,000)
28,000,000
1583
0.060
0.000
0.000
0.000
Volchange
1583
0.011
0.000
0.000
Manchange
1583
0.047
0.000
0.000
Partchange
1583
0.091
0.000
0.000
AWCA
1583
0.339
0.177
0.043
ABHW
1583
0.773
0.013
-0.159
LOSS
1583
0.245
0.000
SIZE
1583
19.112
19.030
LEV
1583
0.539
0.553
SALESGR
1583
1.722
0.067
ROA
1583
3.134
CFO
1583
2.620
0.000
1,050,000,000
Standard Dev.
51,400,000,000
2,920,000,000
0.238
cr
301,000,000
0.000
0.106
0.000
0.213
0.000
0.288
0.335
1.268
0.334
7.193
us
1583
75th Perc
0.000
0.430
17.889
20.351
2.007
0.425
0.667
0.187
-0.036
0.198
39.725
3.500
0.030
7.490
7.959
2.105
1.819
2.381
18.774
an
Net Income
4,110,000,000
25th Perc
1583
Median
Total Assets
Mean
ip
t
Variable
Net Income
984
Change
Median
4,080,000,000
25th Perc
75th Perc
Standard Dev.
415,000,000
180,000,000
1,890,000,000
16,200,000,000
8,725,500
-929,785
50,200,000
777,000,000
134,000,000
Ac
ce
p
Total Assets
Mean
te
Variable
984
0.078
0.000
0.000
0.000
0.269
Volchange
984
0.012
0.000
0.000
0.000
0.110
Manchange
984
0.065
0.000
0.000
0.000
0.247
Partchange
984
0.124
0.000
0.000
0.000
0.330
AWCA
984
0.285
0.176
0.083
0.317
0.578
984
0.385
-0.008
-0.152
0.211
4.535
984
0.240
0.000
0.000
0.000
0.427
984
19.721
19.508
18.522
20.935
1.812
LEV
984
0.551
0.567
0.453
0.664
0.176
SALESGR
984
0.282
0.062
-0.036
0.189
4.465
ROA
984
3.205
3.725
0.210
7.170
7.606
CFO
984
2.116
2.076
1.770
2.353
0.739
ABHW
LOSS
SIZE
39
Page 37 of 49
Count
160,000,000
54,100,000
474,000,000
81,000,000,000
201,000,000
1,645,292
-696,000
9,356,801
4,640,000,000
599
0.030
0.000
0.000
0.000
0.171
Volchange
599
0.010
0.000
0.000
0.000
Manchange
599
0.018
0.000
0.000
0.000
Partchange
599
0.037
0.000
0.000
0.000
AWCA
599
0.795
0.288
0.152
ABHW
599
1.412
0.079
-0.194
LOSS
599
0.254
0.000
0.000
1.000
0.436
SIZE
599
18.113
18.090
17.314
19.188
1.911
LEV
599
0.521
0.524
0.380
0.671
0.204
SALESGR
599
4.087
0.075
-0.031
0.217
64.288
ROA
599
3.018
3.130
-0.330
8.000
8.512
CFO
599
3.447
2.164
1.904
2.416
30.504
ip
t
Change
Standard Dev.
cr
599
75th Perc
0.100
0.134
0.188
0.458
1.801
0.840
10.121
us
Net Income
4,160,000,000
25th Perc
an
599
Median
Total Assets
Mean
Ac
ce
p
te
40
Page 38 of 49
ip
t
cr
TABLE 3
AWCA
us
Correlation matrix
ABHW
LOSS
SIZE
LEV
SALESGR
ROA
CFO
Net Income
0.9709*
Change
0.0137
-0.0016
Volchange
-0.0042
-0.0014
0.4259*
Manchange
0.0179
-0.0007
0.8762*
-0.0208
Partchange
0.0038
0.0027
-0.0603*
-0.0257
-0.0532*
AWCA
-0.0119
-0.0106
-0.0278
-0.0175
-0.019
-0.0327
ABHW
-0.0079
-0.0068
-0.0209
-0.0065
-0.0195
-0.0224
0.0199
LOSS
-0.0268
-0.0369
0.0336*
0.0617*
0.0064
0.0724*
0.0771*
0.0554*
SIZE
0.2194*
0.1587*
0.0875*
0.0078
0.0901*
0.0723*
-0.2044*
-0.1501*
-0.2375*
LEV
0.0101
-0.0121
0.0341
0.0631*
-0.0033
0.0591*
-0.0909*
-0.0397
0.1397*
0.1782*
SALESGR
0.3075*
0.2651*
0.0529*
-0.0077
0.0633*
-0.0205
0.0007
-0.0085
-0.01
0.0932*
0.0165
ROA
0.0266
0.0544*
-0.0129
-0.0226
-0.0022
-0.0416*
-0.0469
-0.0943*
-0.4654*
0.2173*
-0.2398*
-0.0113
CFO
0.4825*
0.4518*
0.0512*
-0.0068
0.0607*
-0.0177
-0.0057
-0.0005
-0.0295
0.1391*
-0.0094
0.7587*
0.0013
BIG
0.0591*
0.0377
-0.0283
-0.0128
0.0027
-0.0728*
0.1759*
-0.0107
0.0165
0.0501*
0.0142
M
an
Total Assets
ce
pt
ed
-0.0741*
-0.002
Ac
-0.0459*
BIG
All variables are defined in Appendix 1. The table reports Pearson correlation coefficients. * denotes significance at the 5% level.
41
Page 39 of 49
ip
t
cr
TABLE 4
Change
(1)Full
AWCA
(2)Big4
AWCA
(3)NonBig4
AWCA
-0.045
(-0.622)
0.057
(0.952)
-0.526**
(-2.117)
Volchange
CFO
LEV
ROA
LOSS
Constant
(6) NonBig4
AWCA
(7)Full
AWCA
(8)Big4
AWCA
(9) NonBig4
AWCA
-0.145
(-0.848)
-0.015
(-0.203)
0.127
(0.763)
0.046
(0.685)
-0.775*
(-1.902)
-0.413**
(-2.084)
0.123
(0.736)
0.042
(0.624)
-0.030
(-0.993)
-0.109***
(-4.208)
-0.023
(-0.849)
-0.339
(-1.498)
0.000
(0.172)
-0.007**
(-2.320)
-0.057
(-1.261)
2.715***
(4.089)
YES
-0.830**
(-2.011)
-0.433**
(-2.140)
-0.820***
(-2.938)
-0.327**
(-2.468)
0.005***
(2.842)
0.181
(0.252)
-0.001**
(-2.293)
-0.003
(-0.278)
0.500
(1.361)
6.248**
(2.288)
YES
984
0.132
599
0.156
-0.109***
(-4.179)
-0.024
(-0.859)
-0.337
(-1.491)
0.000
(0.177)
-0.007**
(-2.314)
-0.059
(-1.287)
2.724***
(4.069)
YES
-0.329**
(-2.479)
0.005***
(2.813)
0.143
(0.199)
-0.001**
(-2.216)
-0.003
(-0.348)
0.469
(1.293)
6.314**
(2.309)
YES
-0.222***
(-3.293)
0.003***
(3.069)
-0.060
(-0.167)
-0.000
(-1.562)
-0.006
(-1.474)
0.081
(0.503)
4.698***
(3.443)
YES
-0.109***
(-4.195)
-0.024
(-0.855)
-0.338
(-1.498)
0.000
(0.194)
-0.007**
(-2.318)
-0.060
(-1.303)
2.721***
(4.077)
YES
-0.330**
(-2.479)
0.005***
(2.812)
0.147
(0.204)
-0.001**
(-2.217)
-0.003
(-0.334)
0.473
(1.302)
6.313**
(2.307)
YES
1,583
0.139
984
0.133
599
0.151
1,583
0.138
984
0.132
599
0.150
1,583
0.140
Ac
SALESGR
(5)Big4
AWCA
-0.221***
(-3.295)
0.003***
(3.068)
-0.062
(-0.173)
-0.000
(-1.554)
-0.006
(-1.479)
0.080
(0.495)
4.692***
(3.443)
YES
ce
pt
SIZE
(4)Full
AWCA
-0.175
(-0.987)
-0.040
(-0.522)
-0.211***
(-2.652)
-0.219***
(-3.277)
0.003***
(3.091)
-0.059
(-0.164)
-0.000*
(-1.697)
-0.006
(-1.456)
0.095
(0.585)
4.647***
(3.422)
YES
Manchange
Partchange
ed
VARIABLES
M
an
us
All Italian listed companies audited by Big 4 and non-Big 4 audit firms. AWCA measure of audit quality. All variables are defined in Appendix 1.
42
Page 40 of 49
ip
t
Ac
ce
pt
ed
M
an
us
cr
43
Page 41 of 49
ip
t
cr
TABLE 5
Change
(1)Full
totrevscaled
(2)Big4
totrevscaled
(3)NonBig4
totrevscaled
-0.001***
(-2.658)
-0.001**
(-2.411)
-0.000
(-0.220)
Volchange
Manchange
SIZE
CFO
LEV
LOSS
Constant
-0.002***
(-3.092)
0.000
(0.541)
-0.004
(-1.253)
0.000
(1.039)
-0.000*
(-1.892)
0.001*
(1.971)
0.037***
(3.142)
YES
1,250
0.090
Ac
SALESGR
ROA
(4)Full
totrevscaled
(5)Big4
totrevscaled
(6) NonBig4
totrevscaled
(7)Full
totrevscaled
(8)Big4
totrevscaled
(9)NonBig4
totrevscaled
-0.002*
(-1.679)
-0.001***
(-2.918)
-0.002
(-1.101)
-0.001***
(-2.701)
-0.000
(-0.420)
0.000
(0.519)
-0.003*
(-1.666)
-0.001***
(-2.777)
-0.001
(-0.603)
-0.002***
(-3.096)
0.000
(0.570)
-0.004
(-1.243)
0.000
(0.965)
-0.000*
(-1.889)
0.001*
(1.925)
0.037***
(3.147)
YES
1,250
0.089
-0.002
(-1.119)
-0.001**
(-2.589)
-0.001
(-0.565)
-0.002***
(-3.020)
0.000
(0.600)
-0.004
(-1.230)
0.000
(0.881)
-0.000*
(-1.887)
0.002*
(1.951)
0.041***
(3.090)
YES
1,109
0.099
-0.000
(-0.393)
0.000
(0.524)
0.000
(0.219)
-0.001***
(-4.127)
0.001
(1.329)
0.003*
(1.893)
-0.001
(-0.815)
0.000
(1.124)
0.000
(0.486)
0.020***
(4.035)
YES
141
0.471
ce
pt
Partchange
ed
VARIABLES
M
an
us
All Italian listed companies audited by Big 4 and non-Big 4 audit firms. Total Audit Fees Scaled by Total Assets as measure of audit fees. All variables are defined in Appendix 1.
-0.002***
(-3.009)
0.000
(0.603)
-0.004
(-1.230)
0.000
(0.888)
-0.000*
(-1.885)
0.002**
(2.015)
0.041***
(3.080)
YES
1,109
0.101
-0.001***
-0.002***
-0.002***
(-4.315)
(-3.083)
(-3.004)
0.001
0.000
0.000
(1.352)
(0.568)
(0.599)
0.002*
-0.004
-0.004
(1.949)
(-1.252)
(-1.236)
-0.001
0.000
0.000
(-0.785)
(0.985)
(0.903)
0.000
-0.000*
-0.000*
(1.109)
(-1.890)
(-1.885)
0.000
0.001**
0.002**
(0.509)
(1.991)
(2.009)
0.020***
0.037***
0.041***
(4.141)
(3.135)
(3.077)
YES
YES
YES
141
1,250
1,109
0.478
0.090
0.100
Standard Error clustered by company.
-0.001***
(-4.125)
0.001
(1.335)
0.003*
(1.904)
-0.001
(-0.792)
0.000
(1.128)
0.000
(0.505)
0.020***
(3.986)
YES
141
0.475
44
Page 42 of 49
ip
t
Ac
ce
pt
ed
M
an
us
cr
45
Page 43 of 49
ip
t
cr
TABLE 6
VARIABLES
Change
M
an
us
All Italian listed companies audited by Big 4 and non-Big 4 audit firms. Total Audit Fees Scaled by Total Assets as measure of audit fees. All variables are defined in Appendix 1.
(1)Full
abnfees
(2)Big4
abnfees
(3)NonBig4
abnfees
-0.001**
(-2.247)
-0.001**
(-2.282)
-0.000
(-0.314)
Manchange
Partchange
CFO
LEV
SALESGR
LOSS
Constant
Ac
ROA
-0.000
(-0.001)
0.000***
(3.158)
0.000
(0.005)
-0.000***
(-3.511)
0.000
(0.002)
0.000
(0.038)
-0.000
(-0.004)
YES
1,250
-0.014
-0.000
(-0.017)
0.000***
(3.073)
-0.000
(-0.139)
-0.000***
(-3.437)
-0.000
(-0.117)
0.000
(0.350)
-0.000
(-0.006)
YES
1,109
-0.016
(5)Big4
abnfees
(6)NonBig4
abnfees
(7)Full
abnfees
(8)Big4
abnfees
(9)NonBig4
abnfees
-0.002
(-1.234)
-0.001***
(-2.933)
-0.002
(-1.111)
-0.001***
(-2.819)
-0.000
(-0.626)
0.000
(0.641)
-0.002
(-1.232)
-0.001***
(-2.800)
-0.001
(-0.583)
0.000
(0.004)
0.000***
(3.091)
0.000
(0.007)
-0.000***
(-3.484)
0.000
(0.004)
0.000
(0.109)
-0.000
(-0.009)
YES
1,250
-0.016
-0.002
(-1.130)
-0.002***
(-2.700)
-0.001
(-0.579)
-0.000
(-0.008)
0.000***
(3.018)
-0.000
(-0.142)
-0.000***
(-3.411)
-0.000
(-0.122)
0.000
(0.386)
-0.000
(-0.014)
YES
1,109
-0.017
-0.000
(-0.584)
0.000
(0.650)
0.000
(0.336)
0.001**
(2.214)
0.000
(0.896)
0.006***
(4.666)
-0.000
(-0.735)
0.000***
(7.656)
-0.001***
(-3.959)
-0.021***
(-3.872)
YES
141
0.676
0.001**
-0.000
-0.000
0.001**
(2.457)
(-0.002)
(-0.015)
(2.224)
0.000
0.000***
0.000***
0.000
(0.965)
(3.112)
(3.041)
(0.910)
0.006***
0.000
-0.000
0.006***
(4.857)
(0.001)
(-0.146)
(4.685)
-0.000
-0.000*** -0.000*** -0.000
(-0.702)
(-3.485)
(-3.410)
(-0.708)
0.000***
0.000
-0.000
0.000***
(7.521)
(0.005)
(-0.117)
(7.605)
-0.001***
0.000
0.000
-0.001***
(-3.793)
(0.059)
(0.358)
(-3.843)
-0.022***
-0.000
-0.000
-0.021***
(-4.170)
(-0.003)
(-0.007)
(-3.854)
YES
YES
YES
YES
141
1,250
1,109
141
0.680
-0.015
-0.017
0.679
Standard Error clustered by company.
Robust t-statistics in parentheses
*** p<0.01, ** p<0.05, * p<0.1
ce
pt
SIZE
ed
Volchange
(4)Full
abnfees
46
Page 44 of 49
ip
t
cr
TABLE 7
VARIABLES
Change
M
an
us
All Italian listed companies audited by Big 4 and non-Big 4 audit firms. Total Audit Fees Scaled by Total Assets as measure of audit fees. All variables are defined in Appendix 1.
(1)Full
ABHW
(2)Big4
ABHW
(3)NonBig4
ABHW
-0.202
(-1.501)
-0.076
(-0.662)
-1.455***
(-2.886)
Manchange
Partchange
CFO
LEV
SALESGR
LOSS
-0.461***
(-2.695)
-0.117
(-0.617)
1.453
(1.617)
-0.010
(-0.795)
0.006
(0.157)
-0.351
(-1.082)
9.005***
(2.684)
YES
1,695
0.032
YES
1,070
0.023
Ac
ROA
-0.605***
(-3.471)
0.012***
(3.111)
0.161
(0.176)
-0.004
(-1.536)
-0.048
(-1.190)
-0.206
(-0.402)
12.436***
(3.661)
Constant
-0.893**
(-2.009)
0.017**
(2.520)
-1.744
(-1.001)
-0.004
(-1.586)
-0.111
(-1.330)
0.233
(0.196)
18.745**
(2.262)
ce
pt
SIZE
ed
Volchange
(4)Full
ABHW
(5)Big4
ABHW
(6)NonBig4
ABHW
(7)Full
ABHW
(8)Big4
ABHW
(9)NonBig4
ABHW
-0.450
(-1.014)
-0.103
(-0.750)
0.200
(0.525)
-0.125
(-1.188)
-2.528**
(-2.436)
-0.789*
(-1.852)
-0.606***
(-3.467)
0.012***
(3.110)
0.167
(0.181)
-0.004
(-1.537)
-0.048
(-1.188)
-0.201
(-0.392)
12.451***
(3.657)
-0.461***
(-2.688)
-0.116
(-0.613)
1.444
(1.602)
-0.010
(-0.793)
0.006
(0.151)
-0.358
(-1.103)
8.999***
(2.680)
-0.895**
(-2.010)
0.017**
(2.521)
-1.725
(-0.989)
-0.004
(-1.586)
-0.112
(-1.329)
0.243
(0.204)
18.767**
(2.261)
-0.495
(-1.102)
-0.144
(-1.041)
-0.384
(-1.256)
-0.600***
(-3.448)
0.012***
(3.098)
0.170
(0.185)
-0.004
(-1.553)
-0.048
(-1.190)
-0.171
(-0.330)
12.369***
(3.643)
0.199
(0.509)
-0.126
(-1.107)
-0.008
(-0.023)
-0.461***
(-2.701)
-0.116
(-0.614)
1.444
(1.603)
-0.010
(-0.787)
0.006
(0.151)
-0.358
(-1.081)
8.998***
(2.681)
-2.657**
(-2.521)
-0.880**
(-2.003)
-2.419***
(-3.453)
-0.889**
(-2.001)
0.017**
(2.533)
-1.626
(-0.927)
-0.004
(-1.645)
-0.111
(-1.323)
0.391
(0.324)
18.666**
(2.253)
YES
1,695
0.031
YES
1,070
0.021
YES
625
0.037
YES
YES
YES
YES
625
1,695
1,070
625
0.038
0.031
0.022
0.036
Standard Error clustered by company.
Robust t-statistics in parentheses
47
Page 45 of 49
ip
t
Ac
ce
pt
ed
M
an
us
cr
48
Page 46 of 49
TABLE 8
Change
Post2change
Post3change
CFO
0.072*
(1.657)
0.074
(1.135)
0.021
(0.530)
0.081
(0.851)
0.230
(1.305)
-0.110***
(-4.246)
-0.023
(-0.828)
-0.367
(-1.541)
0.001
(0.313)
-0.007**
(-2.396)
-0.061
(-1.312)
2.737***
(4.139)
-0.040
(-0.144)
-0.411*
(-1.933)
-0.705**
(-2.329)
-0.701**
(-2.483)
-0.732***
(-2.958)
-0.324**
(-2.468)
0.005***
(2.892)
0.137
(0.193)
-0.001**
(-2.415)
-0.003
(-0.343)
0.534
(1.450)
6.260**
(2.306)
YES
1,583
0.138
YES
984
0.141
YES
599
0.158
Ac
ce
p
te
SALESGR
LEV
LOSS
-0.036
(-0.421)
-0.060
(-0.735)
-0.138*
(-1.694)
-0.133
(-1.237)
-0.027
(-0.175)
-0.220***
(-3.284)
0.003***
(3.045)
-0.049
(-0.134)
-0.000
(-1.519)
-0.006
(-1.443)
0.084
(0.517)
4.733***
(3.348)
SIZE
ROA
(3)NonBig4
AWCA
an
Postchange
(2)Big4
AWCA
cr
Prechange
(1)Full
AWCA
us
VARIABLES
ip
t
All Italian listed companies audited by Big 4 and non-Big 4 audit firms. Prechange, equal to 1 for the year before the change in
audit firm and zero otherwise, Change is equal to 1 for the year of the change in auditor. Postchange, equal to 1 for one year after
the change in audit firm and zero otherwise, Post2change, equal to 1 for the second year after the change in audit firm and zero
otherwise, Post3change, equal to 1 for the third year after the change in audit firm and zero otherwise. All variables are defined in
Appendix 1.
Constant
49
Page 47 of 49
TABLE 9
SIZE
CFO
LEV
SALESGR
ROA
AWCA
AWCA
AWCA
-0.034
(-0.152)
-0.081
(-1.439)
-0.137**
(-2.012)
-0.130***
(-5.072)
-0.002**
(-1.997)
-0.133
(-0.556)
0.002***
(3.205)
-0.005
(-1.322)
0.062
(0.769)
2.641***
(5.534)
0.399
(1.564)
0.031
(0.579)
-0.006
(-0.091)
-0.090***
(-3.415)
0.000
(0.193)
-0.278
(-1.490)
0.000
(1.498)
-0.008*
(-1.684)
-0.028
(-0.455)
1.948***
(3.637)
-0.633**
(-2.150)
-0.311**
(-2.295)
-0.717***
(-4.212)
-0.164***
(-3.524)
-0.001
(-1.399)
-0.062
(-0.142)
0.002***
(6.801)
0.000
(0.002)
0.210
(1.281)
3.392***
(3.765)
0.473*
(1.835)
0.127
(1.237)
0.424 ***
(3.849)
0.370***
(3.498)
0.012***
(4.564)
0.347
(0.758)
-0.007***
(-8.161)
0.012
(1.466)
-0.013
(-0.079)
-7.756***
(-3.929)
0.169
(1.574)
-0.000
(-0.005)
0.169**
(2.408)
0.102***
(2.696)
0.013***
(5.853)
0.344
(1.425)
-0.007***
(-12.835)
0.011*
(1.928)
-0.002
(-0.031)
-2.242***
(-2.870)
1.382***
(2.822)
0.476
(0.958)
0.927**
(2.362)
0.722***
(3.865)
1.038**
(2.024)
0.527
(0.533)
-0.181***
(-2.905)
0.006
(0.322)
-0.521
(-1.260)
-16.904***
(-3.800)
YES
1,102
0.206
YES
745
0.287
YES
357
0.337
Ac
ce
p
Constant
Year Fixed Effects
Observations
Adj. R-squared
YES
1,010
0.164
cr
AWCA
te
LOSS
AWCA
us
Partchange
AWCA
an
Manchange
Volchange
VARIABLES
ip
t
All Italian listed companies audited by Big 4 and non-Big 4 audit firms. (Signed accruals are tested rather than absolute value of
accruals.) All variables are defined in Appendix 1.
YES
620
0.151
YES
390
0.202
50
Page 48 of 49
TABLE 10
(1) AWCA
Volchange
0.038
(0.202)
0.094
(1.075)
-0.080*
(-1.696)
0.324**
(2.159)
-0.643*
(-1.855)
-0.612***
(-2.748)
-0.568***
(-2.969)
-0.197***
(-3.010)
0.003***
(3.075)
-0.088
(-0.253)
-0.001**
(-2.134)
-0.006
(-1.480)
0.102
(0.634)
4.215***
(3.114)
cr
VARIABLES
us
Manchange
Partchange
an
Nonbig4
Volchange*nonbig4
Partchange*nonbig4
SIZE
Manchange*nonbig4
CFO
te
LEV
SALESGR
Ac
ce
p
ROA
ip
t
All Italian listed companies, audited by Big 4 and non-Big 4 audit firms. Non-Big4 binary variable is equal to 1 for firms audited
by a non-Big 4 auditor, and zero otherwise. All variables are defined in Appendix 1.
LOSS
Constant
YES
1,583
0.150
51
Page 49 of 49